Options
2019
Conference Paper
Titel
Randomized Evolution Model for Multi Hypothesis Kalman Filter
Abstract
A new randomized approach for highly maneuvering targets based on multi hypothesis tracking is presented. The acceleration range - a parameter in current evolution models is used to design various motion models. The approach randomises this parameter to cover a wider range of maneuver characteristics. Simulation shows that the performance of the new method results in a more reliable track continuity.